Distinct structural changes occur in the airways of. Matrix Metalloproteinase-9 Expression in Asthma*

Transcription

1 Matrix Metalloproteinase-9 Expression in Asthma* Effect of Asthma Severity, Allergen Challenge, and Inhaled Corticosteroids Waldo Mattos, MD; Sam Lim, MBBS; Richard Russell, MB; Anon Jatakanon, MD; K. Fan Chung, MD, FCCP; and Peter J. Barnes, DM Background: Asthma is associated with remodeling of the extracellular matrix (ECM) and increased airway obstruction, and the mechanisms of this process are unknown. Matrix metalloproteinases (MMPs) are a group of enzymes capable of degrading the ECM. They are released along with their inhibitors, tissue inhibitor of MMP (TIMP). Study objectives: To determine whether severe, persistent asthma is associated with increased levels of MMP-9 in the airway compared with mild asthma, and to assess the effect of both allergen exposure and steroid treatment on MMP-9 and TIMP-1 levels. Design: Prospective analysis of levels and activity of MMP-9 and TIMP-1 in BAL fluid (BALF) and induced sputum obtained from asthmatics of differing disease severity. In patients with mild asthma, MMP-9 and TIMP-1 levels were studied in induced sputum following allergen challenge and in BALF after inhaled steroid therapy. Patients: Eighteen patients with mild asthma, 10 patients with severe asthma, and 10 nonsmoking, atopic subjects had their sputum studied. Fourteen of the patients with mild asthma underwent allergen challenge. BAL was collected from 16 patients with mild asthma before and after 4 weeks treatment with inhaled budesonide, 800 g bid, or placebo. Results: Patients with severe asthma had increased levels and activity of sputum MMP-9 in their sputum compared with patients with mild asthma and normal subjects. Allergen challenge increased the MMP-9/TIMP-1 ratio and MMP-9 activity. Inhaled budesonide had no effect on MMP-9 or TIMP-1 in patients with mild asthma. Conclusions: MMP-9 may play a role in chronic airway inflammation and remodeling in asthma, as concentrations are increased in severe, persistent asthma and following allergen challenge. Inhaled steroids may not affect MMP-9 and TIMP in patients with mild asthma, and additional studies in patients with more severe asthma are needed. (CHEST 2002; 122: ) Key words: allergen challenge; asthma; corticosteroid; matrix metalloproteinase; matrix metalloproteinase-9; tissue inhibitors of matrix metalloproteinase Abbreviations: BALF BAL fluid; CI confidence interval; ECM extracellular matrix; IL interleukin; MMP matrix metalloproteinase; PC 20 20% fall in FEV 1 ; TIMP tissue inhibitor of matrix metalloproteinase Distinct structural changes occur in the airways of asthmatic patients, including fibrosis under the epithelium and within the airway wall. 1 4 However, there is a surprising lack of fibrosis or connective tissue destruction in asthma compared with other *From the Department of Thoracic Medicine, National Heart and Lung Institute, Faculty of Medicine, Imperial College, London, UK. Dr. Mattos was supported by a fellowship from CAPES, Ministry of Education, Brazil. Manuscript received September 27, 2001; revision accepted May 24, Correspondence to: Peter J. Barnes, DM, Department of Thoracic Medicine, National Heart and Lung Institute, Dovehouse St, London WS3 6LY, UK; p.j.barnes@ic.ac.uk chronic inflammatory or obstructive diseases of the airway that involve neutrophils, suggesting the existence of protective mechanisms against fibrosis and reversal of the fibrosis in the eosinophilic inflammation of asthma. The basement membrane is a thin layer of specialized extracellular matrix (ECM) beneath the bronchial epithelium, which is composed of a mixture of collagens, elastin, proteoglycans, laminin, and fibronectin. It provides structural support and spatial orientation to overlying cells, and also regulates macromolecule and cell traffic. 5 As part of the inflammatory response in asthma, proteolytic enzymes, such as the matrix metalloproteinases CHEST / 122 / 5/ NOVEMBER,

2 (MMPs) are continually degrading the structure of ECM. These enzymes may facilitate leukocyte migration through the ECM and between endothelial cells, 6 8 and also affect other aspects of cell activation and survival in the tissues by cleaving cytokines and their receptors from the cell surface. 9,10 The turnover and remodeling of ECM is highly regulated, since either uncontrolled proteolysis or excess collagen deposition may be associated with structural changes. The MMPs are a large family of zinc- and calciumdependent enzymes produced by a wide range of stromal and inflammatory cells, and are capable of degrading all of the components of the ECM. 11,12 Most members of the MMP family are organized into three structural domains: an amino-terminal propeptide, a catalytic domain, and a hemopexin-like domain. 12 The subfamilies MMP-2 and MMP-9, also known as gelatinases, have incorporated a gelatinbinding domain, rendering the structure of the gelatinases unique. 13 In common with all MMPs, the gelatinases are produced in a latent form requiring activation to produce their enzymatic effects, and can be inactivated when combined with specific inhibitors, tissue inhibitor of metalloproteinase (TIMP). 14 There is indirect evidence for a role for MMPs in the pathology of asthma. MMP-9 (gelatinase B) is highly expressed in bronchial biopsies from asthmatic patients compared with normal subjects. 15 Other studies have demonstrated increased levels of MMP-9 in BAL fluid (BALF) and alveolar macrophage supernatant in untreated asthmatics, 16 and increased MMP-9 activity measured by zymography in BALF of asthmatic patients after allergen challenge. 17 Furthermore, Vignola and coworkers 18 reported increased sputum levels of both MMP-9 and TIMP-1, but a decreased MMP-9/TIMP-1 ratio, in both asthmatic and chronic bronchitic patients compared with normal control subjects. MMP-9 expression may increase during inflammation of the airways in asthmatics, but it is not certain whether this has any effect on airway remodeling or whether inhaled corticosteroid treatment has any effect on the expression of MMP-9 or its major inhibitor, TIMP-1, in vivo. The effect of inhaled steroids on ECM structure and, more specifically, on basement membrane thickening is not well established. Jeffery et al 19 demonstrated in a small number of patients that neither short- nor long-term treatment with inhaled budesonide reduced the thickening of the reticular basement membrane. By contrast, others have reported a reduction in the thickness of type III collagen deposition in the bronchial lamina reticularis after more prolonged treatment with an inhaled steroid, 20 and another study reported that treatment with higher doses of inhaled steroids slowed basement membrane thickening. 21 Although inhaled steroids are able to reduce various inflammatory parameters in asthma, 19,20 and steroids potently inhibit MMP-9 synthesis in vitro, 22 their ability to interfere with either the proteolytic activity of MMPs or the regulation of the inhibitory mechanisms in vivo is not yet certain. Corticosteroids inhibit the expression of multiple proinflammatory cytokines, such as macrophage inhibitory protein-1, granulocyte-macrophage colony stimulating factor, and tumor necrosis factor-, and enhance anti-inflammatory cytokines such as interleukin (IL) MMP-9 expression may be increased by these inflammatory cytokines. 24 Obtaining specimens from subjects with respiratory disease can be problematic. Bronchoscopy is invasive and not without risk. Sputum induction is a less invasive method for obtaining samples for study, which is safe and repeatable. We therefore undertook a study to compare sputum MMP-9 and TIMP-1 between normal subjects and asthmatic patients with differing disease severity and to also study the effect of allergen exposure, which induces acute airway inflammation on the sputum levels of MMP-9 and TIMP-1 and TIMP-2 in patients with mild asthma. The effect of inhaled budesonide on MMP-9 and TIMP-1 was studied in BALF from asthmatic patients and normal control subjects. Patient Selection Materials and Methods The diagnosis of bronchial asthma was based on the criteria of the American Thoracic Society. 25 Asthmatic patients demonstrated airway hyperresponsiveness with a provocative concentration of methacholine producing a 20% fall in FEV 1 (PC 20 )of 4 mg/ml, and were atopic as defined by two or more positive skin-prick test results to common aeroallergens. Current smokers or ex-smokers of more than five pack-years were excluded. All patients gave informed written consent to these studies, which were approved by the Royal Brompton Hospital Ethics Committee. Induced Sputum Study Ten nonsmoking, nonatopic, healthy subjects; 20 patients with mild, stable asthma; and 10 patients with severe asthma were recruited to the study (Table 1). The patients with asthma were receiving treatment only with inhaled short-acting 2 -adrenergic agonist aerosol for intermittent relief of wheeze. The patients with severe asthma were receiving regular treatment with longacting bronchodilators and high doses of inhaled corticosteroids, and some patients were treated with low-dose oral corticosteroids. BAL Study Eight nonsmoking, nonatopic, healthy subjects (average age, 25.5 years) and 16 patients with mild, stable asthma (Table 1), 1544 Clinical Investigations

3 Characteristics Table 1 Patient Characteristics* Sputum Study Healthy Subjects Mild Asthma Severe Asthma BAL Study Mild Asthma Subjects, No Age, yr 32.8 ( ) 26.1 (24 28) 44.4 ( ) 24 ( ) Male/female 4/6 7/11 3/7 9/7 gender, No. FEV 1, % predicted ( ) 94.1 ( ) 53.4 ( ) 92.1 ( ) PC 20, mg/ml ( ) ND 1.6 ( ) Oral steroid Inhaled steroid *Data are shown as mean (95% CI) unless otherwise indicated. ND not done. Prednisolone, 10 to 15 mg/d. Budesonide or fluticasone propionate, 1,600 to 2,000 g/d. who were receiving treatment only with the inhaled 2 -adrenergic agonist aerosol for intermittent relief of wheeze, were recruited. Study Design Induced Sputum Study: We studied MMP-9 and TIMP-1 concentrations and MMP-9 activity in sputum supernatant from nonasthmatic normal subjects and patients with mild and severe asthma. MMP-9 levels in saliva, collected immediately before the sputum induction, were compared with those in sputum in the control group. At the screening visit, between 8 am and 12 am, subjects underwent spirometry, skin-prick testing, and methacholine challenge, followed by sputum induction. In order to investigate the effect of allergen exposure, we studied the sputum concentrations of MMP-9, TIMP-1, and MMP-9 activity 5 to 7 days before and 24 h after allergen challenge in 14 patients with mild asthma who had previously demonstrated a late-phase reaction to inhaled allergen. BAL Study: After initial screening with spirometry, skin-prick testing, methacholine challenge, fiberoptic bronchoscopy, and BAL was performed in healthy and asthmatic subjects. We studied the MMP-9 and TIMP-1 concentrations in BAL from 8 healthy subjects and 16 patients with mild asthma. In order to investigate the effect of inhaled steroid treatment on the BALF levels of MMP-9 and TIMP-1, the asthmatic patients were then randomized to receive treatment with budesonide (800 g bid via a Turbuhaler [Astra Zeneca; Lund, Sweden]) or matched placebo for 4 weeks, in a double-blind, parallel-group study. Compliance was evaluated by counting the doses remaining. PC 20 and BAL were repeated after the 4-week treatment period. Lung Function and Methacholine Challenge FEV 1, FVC, and methacholine challenge were performed at the screening visit. Doses of methacholine were administered starting at a dose of 0.06 mg/ml via a Mefar dosimeter (Mefar; Bovezzo, Italy). The PC 20 was calculated from concentrationresponse curves by linear interpolation. Allergen Challenge and Skin-Prick Testing Freeze-dried allergen extracts (Aquagen SQ; Allergologisk Laboratium; Horsholm, Denmark) were used. The allergen giving the strongest skin-prick response (house dust mite or cat allergen only) was selected for airway challenge. Known dilutions of the allergen were prepared to give final concentrations of 250, 500, 1,000, 2,000, 4,000, 8,000, 16,000, and 32,000 squamous epithelial cells per milliliter. Allergen aerosols were delivered from a nebulizer attached to a breath-activated dosimeter (Mefar). When a fall in FEV 1 of 15% was recorded, the challenge was stopped and the FEV 1 measured after 5, 10, 20, 30, 45, 60 min and thereafter every 30 min up to 10 h. The late-phase response was defined as the presence of at least three FEV 1 measurements after the fourth hour that were at least 15% below the baseline FEV 1 value. Patients were then allowed to go home but returned to the laboratory 24 h after the start of the allergen challenge. Sputum Induction and Processing Sputum was induced and processed using the method described by Keatings et al. 26 The ultrasonic nebulizer (DeVilbiss; Heston, UK) was utilized at maximum output (4 ml/min) for 15 min. The initial sample from the first cough was discarded. Sputum was collected into a 50-mL polypropylene tube, kept at 4 C, and processed within 2 h. All slides were blinded before counting, and 300 inflammatory cells were counted in each sample. An adequate sample was defined as having 50% of squamous epithelial cells on cytospin. Samples of saliva, collected in a polypropylene tube, were processed in exactly the same way described for sputum. All supernatants were stored at 70 C for further analysis. Fiberoptic Bronchoscopy BAL was performed in the right middle lobe using four successive aliquots of 60 ml of 0.9% NaCl. BALF was spun at 500g for 10 min, and the supernatant was separated and concentrated 10 times using a microconcentrator (3K Microsep; Flowgen Instruments; Maidstone, UK) and stored at 70 C for further analysis. Enzyme-Linked Immunosorbent Assay MMPs or TIMPs were assayed in the sputum or saliva supernatant using a quantitative sandwich enzyme immunoassay technique (Biotrak; Amersham Pharmacia Biotech; Amersham, UK). The sensitivity of the assays were 0.6 ng/ml for MMP-9 and 1.25 ng/ml for TIMP-1. MMP assays recognize latent and active MMP-9 and that complexed with TIMP. TIMP antibodies bind to free and complexed TIMP-1. All results were corrected for the volume and dilution of sputum or saliva. CHEST / 122 / 5/ NOVEMBER,

4 Zymography for MMP-9 Activity Sputum samples were incubated overnight at 37 C with 30 U of galactosidase to lyse long-chain polysaccharides, which prevent samples running in the gel; 7.5% polyacrylamide sodium dodecylsulfate gels incorporating 0.1% gelatin were made, and 30 g of protein from samples were loaded to each well of the plate and run for 3 h at 40 ma per gel. Gels were incubated overnight (37 C) in Tris buffer containing 5 mol/l zinc chloride and 10 mm calcium chloride. Gels were stained with Coomassie blue, and the results were read using image-analysis software (Gel Works; UVP; Cambridge, UK). Areas of enzymatic activity are seen as clear against the stained background. Activity was expressed as relative absorbance as a percentage of activity of known amounts of control MMP in order to correct for gel-to-gel variations. BAL samples did not require pretreatment with galactosidase and were loaded by volume (15 L). Characteristics Table 2 Sputum Characteristics* Healthy Subjects Mild Asthma Severe Asthma Subjects, No Volume, ml 3.7 ( ) 2.4 ( ) 2.5 ( ) SQU, % 32.7 ( ) 10 (5 47.5) 29.4 ( ) TIC, 10 6 /ml 2.1 ( ) 3.4 ( ) 2.4 ( ) Macrophages, % 59.7 ( ) 62.1 ( ) 49.2 ( ) Neutrophils, % 37.7 ( ) 33.3 ( ) 29.3 ( ) Eosinophils, % 0.1 (0 0.7) 1.4 ( ) 4.0 ( ) Lymphocytes, % 1.9 ( ) 0(0 0.4) 1.6 ( ) *Data are shown as median (25 to 75 percentiles); SQU squamous epithelial cells; TIC total inflammatory cell count. p 0.05 compared with control. p compared with control. p compared to patients with mild asthma. BAL Albumin Levels Albumin levels in BALF were quantified by rocket immunoelectrophoresis using specific antisera and standardized against standard human serum containing albumin of known concentration. All values were expressed in nanograms per milligram of albumin measured in BALF (milligrams per liter). Albumin antisera was obtained from DAKO Ltd. (High Wycombe, UK), and the standard human serum containing albumin of known concentration was obtained from Behring Ltd. (Hounslow, UK). Data Analysis FEV 1 and PC 20 were analyzed using paired Student t tests in the comparisons of data before and after allergen challenge. Results of parametric data are expressed as means with 95% confidence intervals (CIs) unless stated otherwise. Cell count data and results of supernatant assays are not normally distributed and are expressed as medians and percentiles. The Kruskal- Wallis test was used to compare data from control subjects and patients with mild and severe asthma. The Wilcoxon signed-rank test was used to compare data before and after allergen exposure and inhaled steroid therapy, and the Mann-Whitney rank-sum test was used to compare different groups. These results are expressed as medians and 25 to 75% percentiles. For all tests, p 0.05 was considered significant. 76.4%). There was no significant change in PC h after the allergen challenge (Table 3). There was an increase in the percentage of eosinophils after allergen challenge (p 0.001; Table 3). Sputum MMP-9 and TIMP-1 Levels MMP-9 was only detected in the saliva of six subjects. The concentration of MMP-9 measured was much lower than that in sputum (median, 34 times lower). We did not observe any difference in sputum MMP-9 levels between healthy subjects and patients with mild asthma at baseline. However, patients with severe asthma had significantly higher concentrations of MMP-9 compared to both healthy subjects and patients with mild asthma (p and p 0.01, respectively). There was an increased level of TIMP-1 (p 0.05) and a tendency to higher MMP-9/TIMP-1 ratio (p 0.06) in patients with severe asthma compared to patients with mild asthma (Fig 1). There was no correlation between the Results Inflammatory Cells in Induced Sputum and Effects of Allergen Challenge Two subjects with mild asthma were excluded, as their sputum samples showed 50% squamous cells. Patients with both mild and severe asthma had an increased percentage of eosinophils compared with the control group (p 0.05 and p , respectively; Table 2). There was a small but statistically significant higher percentage of lymphocytes in the control group (p 0.001). During the early-phase reaction of the allergen challenge, FEV 1 dropped to 76.2% of the postsaline solution value (95% CI, 71.5 to 80.8%), and during late-phase reaction to 69.6% (95% CI, 62.7 to Table 3 Effects of Allergen Exposure* Variables Baseline 24 h After Subjects, No FEV 1, % predicted 95.7 ( ) 90.5 ( ) PC 20, mg/ml 1.1 ( ) 0.8 ( ) Sputum characteristics Volume, ml 3.2 (2 4) 2.7 ( ) SQU, % 7.5 (5 32.5) 15 ( ) TIC, 10 6 /ml 3.5 ( ) 3.5 ( ) Macrophages, % 59.6 ( ) 43 ( ) Neutrophils, % 35.7 ( ) 47.7 ( ) Eosinophils, % 1.3 ( ) 8.5 ( ) Lymphocytes, % 0 (0 0.4) 0.1 (0 0.2) *FEV 1 and PC 20 values are shown as mean (95% CI). Sputum characteristic values are shown as median (25 to 75 percentiles). See Table 2 for expansion of abbreviations. p Clinical Investigations

5 concentrations of MMP-9 or TIMP-1 and the percentage of inflammatory cells, including eosinophils, in sputum. In patients with a late-phase reaction, there was a small but statistically significant increase in the MMP-9/TIMP-1 ratio after allergen exposure (p 0.05; Fig 2, bottom, C). MMP-9 Activity MMP-9 activity was increased in the sputum supernatant from patients with severe asthma when compared to patients with mild asthma and normal subjects, as seen by gelatinolytic activity at 92 kd: Figure 1. MMP-9 (top, A), TIMP-1 (center, B), and MMP-9/ TIMP-1 ratio (bottom, C) in induced sputum from healthy subjects and patients with mild and severe asthma. Values are shown as individual data points with median values. *p 0.05; **p 0.01; ***p Figure 2. Effect of allergen challenge on MMPs and TIMPs in sputum of patients with mild asthma. Measurements of MMP-9 (top, A), TIMP-1 (center, B), and MMP-9/TIMP-1 ratio (bottom, C) before and 24 h after allergen exposure are shown. Values are shown as individual data points with median values. *p CHEST / 122 / 5/ NOVEMBER,

6 Effect of Budesonide on FEV 1 and Bronchial Responsiveness No difference was found in the baseline FEV 1 (budesonide group, 92.3% of predicted value [95% CI, 86.0 to 98.6%]; placebo group, 92.0%, [95% CI, 85.0 to 99.0%]) and baseline PC 20 between any of the groups (budesonide group, 0.95 mg/ml [95% CI, 0 to 2.21 mg/ml]; placebo group, 2.59 mg/ml [95% CI, 1.26 to 3.93 mg/ml]). The compliance with the treatment was 80% of the doses, as measured by doses remaining in the Turbuhaler. There was no difference in the FEV 1 4 weeks after placebo or budesonide, but there was a significant increase in PC 20 after budesonide (p 0.05). MMP-9 and TIMP-1 BALF Levels and Activity, Effects of Budesonide Treatment Figure 3. Top, A: Zymographic analysis of MMP-9 in sputum in all groups (values are shown as individual data points and means [SE]). *p ; **p 0.01; ***p Center, B: Zymographic analysis of patients with mild asthma before and after allergen exposure (values are shown as individual data points with median values). Bottom, C: Gelatinolytic activity at 92 kd (MMP-9) and 72 kd (MMP-2) of sputum sample of patients with severe asthma. 67.5% (95% CI, 48.0 to 87.0%) vs 35.4% (95% CI, 27.8 to 43.1) and 20.6% (95% CI, 12.0 to 29.2%), respectively (p 0.001; Fig 3, top, A). There was also an increased activity at 82 kd, which represents activated MMP-9 in patients with severe asthma compared to patients with mild asthma and normal subjects (that is in the fully active form with a 10-kd propeptide removed). No significant difference was seen between normal subjects and patients with mild asthma. After allergen challenge, there was a significant increase in MMP-9 activity: 31.6% (95% CI, 22.9 to 46.4%) vs 41.2% (95% CI, 34.0 to 48.8%) [p 0.002; Fig 3, center, B]. The baseline concentration of albumin in BALF was not statistically different between each group: control (median, 24 mg/l; percentiles, 16 to 28.7 mg/l), asthmatics receiving budesonide (median, 33 mg/l; percentiles, 29 to 35 mg/l), and asthmatics receiving placebo (median, 26 mg/l; percentiles, 24 to 41.5 mg/l). In asthmatic patients, albumin levels were not affected by steroid (median, 32 mg/l; percentiles, 21 to 35 mg/l) or placebo (median, 47 mg/l; percentiles, 30 to 63.5 mg/l). There was a significant higher baseline concentration of TIMP-1 in asthmatic patients compared with normal subjects (p 0.05; Fig 4, bottom, B), but there was no difference within the asthmatic group before and after treatment with budesonide or placebo (Fig 5). There was no difference in MMP-9 level either between asthmatic and normal subjects, and within the asthmatic group before and after treatment with budesonide or placebo (Figs 4, 5). There was no significant difference in the MMP- 9/TIMP-1 ratio between asthmatic patients (median, 0.015; percentiles, 0 to 0.072) and control subjects (median, 0.052; percentiles, 0 to 0.28), either in asthmatic subjects after inhaled corticosteroid treatment (before budesonide median, [percentiles, to 0.09]; after budesonide median, [percentiles, 0 to 0.07]) or placebo (before placebo median, [percentiles, 0 to 0.006]; after placebo median, [percentiles, 0.01 to 0.04]). Zymographic analysis revealed no difference in MMP-9 activity after treatment with budesonide in asthmatic subjects (Fig 5, bottom right, F). We did not find any increase in eosinophil numbers in BALF at baseline, and it was not possible to correlate this cell with MMP-9 and TIMP Clinical Investigations

7 Figure 4. Comparison of MMP-9 (top, A) and TIMP-1 (bottom, B) concentrations in BALF between control subjects and asthmatic patients (values are shown as individual data points and medians). *p Discussion We have demonstrated in this study that the sputum concentration of MMP-9 is increased in patients with severe asthma compared with either healthy subjects or patients with mild asthma. Furthermore, the MMP-9/TIMP-1 ratio was higher in patients with severe asthmatic. In parallel, we showed that the sputum contamination with saliva does not interfere with MMP-9 levels in sputum. We also demonstrated an increased gelatinolytic activity in induced sputum from asthmatic patients compared with control subjects and that activity increases following allergen challenge. Indeed, the MMP-9/TIMP-1 ratio increased following this stimulus. These findings may indirectly indicate the involvement of MMP-9 in airway remodeling process in patients with severe asthma who have an irreversible component of airway narrowing (with a mean FEV 1 of approximately 50% predicted normal) and in those who, due to environmental factors, have a high allergen load. MMP-9 may play a role in tissue remodeling during physiologic and pathologic processes by initiating the degradation of the ECM, and MMP overexpression may lead to tissue destruction that is characteristic of chronic inflammatory diseases such as rheumatoid arthritis and scleritis. 27 There is some evidence supporting an increase in ECM degradation in asthma. Structural changes in the cartilage in airways of 1 to 5 mm in diameter were described in six fatal asthma cases, which may be indicative of a more extensive destructive process in the airways. 28 Furthermore, Lemjabbar et al 29 showed an acute 10-fold to 160-fold increase of MMP-9 and MMP-3 concentration in the epithelial lining fluid of patients undergoing mechanical ventilation for status asthmaticus. Another possible role of MMP-9 in asthma is its place in enabling cell migration and airway inflammatory response. Gelatinases, especially MMP-9, are up-regulated in vitro by a group of predominately proinflammatory cytokines, 30,31 and some increase in proteolytic activity might be expected during acute airway eosinophilic inflammation. In vitro 32 and in vivo 7 studies suggest that MMP-9 may play an important role during eosinophil migration. Tumor necrosis factor- and IL-1 selectively induce the expression of MMP-9 in human macrophages, 33 whereas IL-4 and IL-10 inhibit MMP production monocyte/macrophages. 34,35 It is therefore possible that the balance of different cytokines may determine the production of MMPs. We also demonstrated that in patients with mild, stable asthma, there is a higher concentration of TIMP-1 in BALF than healthy, nonsmoking subjects, despite similar levels of MMP-9. It is possible that the increased levels of TIMP-1 observed in asthmatics may represent an endogenous protective mechanism to down-regulate the proteolytic activity of MMPs in lung parenchyma or that an excess of TIMP-1 could lead to airway fibrosis. 36 Inhaled corticosteroids are the mainstay of modern asthma management. We have also shown that 4 weeks of treatment with a high dose of corticosteroid did not produce any change in the BALF concentrations of MMP-9 or TIMP-1 or of MMP-9 activity. In patients with severe asthma, there is a positive correlation between FEV 1 improvement with corticosteroid treatment and serum MMP-9/ TIMP-1 ratio, 37 suggesting that in patients with a low MMP-9/TIMP-1 ratio and little response with corticosteroids, bronchial fibrogenesis predominates over inflammation. In vitro studies have shown that steroids down-regulate the production of TIMP-1 and MMP-9 from human alveolar macrophage. 16,22 In our study, inhaled steroid decreases airway responsiveness but did not affect MMP-9 activity or protein levels. However, our subjects had mild asthma and thus only had minor airway inflammatory changes present. It follows that inhaled steroids might be expected to have little effect, as baseline inflammatory activity should be low. Furthermore, BAL sam- CHEST / 122 / 5/ NOVEMBER,

8 Figure 5. MMP-9 concentration in BALF of control subjects and asthmatic patients before and after treatment with placebo (top left, A) and budesonide (top right, B); TIMP-1 concentration in BALF of control subjects and asthmatic patients before and after treatment with placebo (center left, C) and budesonide (center right, D); MMP-9 activity in BALF of control subjects and asthmatic patients before and after treatment with placebo (bottom left, E) and budesonide (bottom right, F). Values are shown as individual data points. ples may be more representative of the lung parenchyma than the airways, as asthma is predominantly an airway disease. The source of MMP-9 and TIMP in asthma is at present unclear. Although it has been demonstrated by Ohno et al 15 that eosinophils are an important source of MMP-9 in asthma, other cells such as neutrophils and alveolar macrophages also release MMP-9. A previous study has reported immunoreactivity for MMP-3 and MMP-9 in endobronchial biopsy specimens from patients with asthma of differing severity, and MMP-9 was prominent within the ECM as well as in neutrophils. 38 In our study, there was no significant increase in neutrophils in the group of patients with severe asthma compared to patients with mild asthma and control subjects. The fact that neutrophils were not significantly increased in the patients with severe asthma may reflect the fact that this group of patients was characterized by an increase in eosinophils and the numbers of patients in this group was relatively small. There was no significant correlation between MMP-9 levels or activity and sputum inflammatory cells, although macrophages are the predominant cell. The lack of correlation with particular types of 1550 Clinical Investigations

9 inflammatory cell may reflect the fact that MMP-9 is derived from several cell types, even including structural cells in the airway wall. It is possible that due to the heterogeneity of asthma, MMP-9 may play a role in the pathogenesis of a group of patients with more severe asthma, particularly those with nonreversible airway obstruction. The results of our study suggest that MMP-9 may play a role in the airway remodeling in both the chronic inflammatory process in patients with severe asthma, particularly those with nonreversible obstruction, and acute eosinophilic inflammatory response in mild asthma. However, inhaled steroids in patients with mild asthma did not change MMP-9 expression or activity. The mechanisms of regulation of this enzymatic system and the cellular source of MMPs need to be clarified in future studies. References 1 Laitinen LA, Heino M, Laitinen A, et al. Damage of the airway epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 1985; 131: Jeffery PK, Wardlaw AJ, Nelson FC, et al. Bronchial biopsies in asthma: an ultrastructural, quantitative study and correlation with hyperreactivity. Am Rev Respir Dis 1989; 140: Roche WR, Beasley R, Williams JH, et al. Subepithelial fibrosis in the bronchi of asthmatics. Lancet 1989; 1: Redington AE, Sime PJ, Howarth PH, et al. Fibroblasts and the extracellular matrix in asthma. In: Marone G, Austen KF, Holgate ST, et al, eds. Asthma and allergic diseases: physiology, immunopharmacology, and treatment. London, UK: Academic Press, 1998; Xia M, Leppert D, Hauser SL, et al. Stimulus specificity of matrix metalloproteinase dependence of human T cell migration through a model basement membrane. J Immunol; 1996; 156: Leppert D, Waubant E, Galardy R, et al. T cell gelatinases mediate basement membrane transmigration in vitro. J Immunol 1995; 154: Delclaux C, Delacourt C, D Ortho MP, et al. Role of gelatinase B and elastase in human polymorphonuclear neutrophil migration across basement membrane. Am J Respir Cell Mol Biol 1996; 14: Gearing AJ, Beckett P, Christodoulou M, et al. Processing of tumour necrosis factor- precursor by metalloproteinases. Nature 1994; 370: Kayagaki N, Kawasaki A, Ebata T, et al. Metalloproteinasemediated release of human Fas ligand. J Exp Med 1995; 182: Murphy G, Docherty AJ. The matrix metalloproteinases and their inhibitors. Am J Respir Cell Mol Biol 1992; 7: O Connor CM, FitzGerald MX. Matrix metalloproteinases and lung disease. Thorax 1994; 49: Massova I, Kotra LP, Fridman R, et al. Matrix metalloproteinases: structures, evolution, and diversification. FASEB J 1998; 12: Murphy G, Nguyen Q, Cockett MI, et al. Assessment of the role of the fibronectin-like domain of gelatinase A by analysis of a deletion mutant. J Biol Chem 1994; 269: Okada Y, Gonoji Y, Naka K, et al. Matrix metalloproteinase 9 (92-kd gelatinase/type IV collagenase) from HT 1080 human fibrosarcoma cells: purification and activation of the precursor and enzymic properties. J Biol Chem 1992; 25: Ohno I, Ohtani H, Nitta Y, et al. Eosinophils as a source of matrix metalloproteinase-9 in asthmatic airway inflammation. Am J Respir Cell Mol Biol 1997; 16: Mautino G, Oliver N, Chanez P, et al. Increased release of matrix metalloproteinase-9 in bronchoalveolar lavage fluid and by alveolar macrophages of asthmatics. Am J Respir Cell Mol Biol 1997; 17: Warner JA, Julius P, Luttmann W, et al. Matrix metalloproteinases in bronchoalveolar lavage fluid following antigen challenge. Int Arch Allergy Immunol 1997; 113: Vignola AM, Riccobono L, Mirabella A, et al. Sputum metalloproteinase-9/tissue inhibitor of metalloproteinase-1 ratio correlates with airflow obstruction in asthma and chronic bronchitis. Am J Respir Crit Care Med 1998; 158: Jeffery PK, Godfrey RW, Adelroth E, et al. Effects of treatment on airway inflammation and thickening of basement membrane reticular collagen in asthma. Am Rev Respir Dis 1992; 145: Trigg CJ, Manolitsas ND, Wang J, et al. Placebo-controlled immunopathologic study of four months of inhaled corticosteroids in asthma. Am J Respir Crit Care Med 1994; 150: Sont JK, Willems LN, Bel EH, et al. Clinical control and histopathologic outcome of asthma when using airway hyperresponsiveness as an additional guide to long-term treatment. Am J Respir Crit Care Med 1999; 159: Hoshino M, Takahashi M, Takai Y, et al. Inhaled corticosteroids decrease subepithelial collagen deposition by modulation of the balance between matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 expression in asthma. J Allergy Clin Immunol 1999; 104: John M, Lim S, Seybold J, et al. Inhaled corticosteroids increase interleukin-10 but reduce macrophage inflammatory protein-1, granulocyte-macrophage colony-stimulating factor, and interferon- release from alveolar macrophages in asthma. Am J Respir Crit Care Med 1998; 157: Zhang Y, McCluskey K, Fujii K, et al. Differential regulation of monocyte matrix metalloproteinase and TIMP-1 production by TNF-, granulocyte-macrophage-csf, and IL-1- through prostaglandin-dependent and independent mechanisms. J Immunol 1998; 161: American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 1987; 136: Keatings VM, Jatakanon A, Worsdell YM, et al. Effects of inhaled and oral glucocorticosteroids on inflammatory indices in asthma and COPD. Am J Respir Crit Care Med 1997; 155: Di Girolamo N, Tedla N, Lloyd A, et al. Expression of matrix metalloproteinases by human plasma cells and B lymphocytes. Eur J Immunol 1998; 28: Roberts CR, Okazawa M, Wiggs B, et al. Airway wall thickening. In: Barnes PJ, Grunstein MM, Leff AR, et al, eds. Asthma. 1st ed. Philadelphia, PA: Lippincott-Raven Publishers, 1997; Lemjabbar H, Gosset P, Lamblin C, et al. Contribution of 92 kd gelatinase/type IV collagenase in bronchial inflammation during status asthmaticus. Am J Respir Crit Care Med 1999; 159: Yao PM, Maitre B, Delacourt C, et al. Divergent regulation of 92 kd gelatinase and TIMP-1 by HBECs in response to IL-1 and TNF-. Am J Physiol 1997; 273: Johnatty RN, Taub DD, Reeder SP, et al. Cytokine and chemokine regulation of prommp-9 and TIMP-1 production CHEST / 122 / 5/ NOVEMBER,

10 by human peripheral blood lymphocytes. J Immunol 1997; 158: Okada S, Kita H, Georga TJ, et al. Migration of eosinophils through basement membrane components in vitro: role of matrix metalloproteinase-9. Am J Respir Cell Mol Biol 1997; 17: Saren P, Welgus HG, Kovanen PT. TNF- and IL-1 selectively induce expression of 92-kd gelatinase by human macrophages. J Immunol 1996; 157: Corcoran ML, Stetler SW, Brown PD, et al. Interleukin 4 inhibition of prostaglandin E2 synthesis blocks interstitial collagenase and 92-kd type IV collagenase/gelatinase production by human monocytes. J Biol Chem 1992; 267: Mertz PM, DeWitt DL, Stetler SW, et al. Interleukin 10 suppression of monocyte prostaglandin H synthase-2: mechanism of inhibition of prostaglandin-dependent matrix metalloproteinase production. J Biol Chem 1994; 269: Mautino G, Henriquet C, Jaffuel D, et al. Tissue inhibitor of metalloproteinase-1 levels in bronchoalveolar lavage fluid from asthmatic subjects. Am J Respir Crit Care Med 1999; 160: Bosse M, Chakir J, Rouabhia M, et al. Serum matrix metalloproteinase-9: tissue inhibitor of metalloproteinase-1 ratio correlates with steroid responsiveness in moderate to severe asthma. Am J Respir Crit Care Med 1999; 159: Dahlen B, Shute J, Howarth P. Immunohistochemical localisation of the matrix metalloproteinases MMP-3 and MMP-9 within the airways in asthma. Thorax 1999; 54: Clinical Investigations